We are using 1760 routers with a dual t1 card, and occasionally an unknown event (possibly within some microwave equipment) will trigger two of the routers’ t1 lines to fail.
Router A fails with a full set of slip seconds (900 per 15 minutes) and Router B fails with a large number of line code and path code violations on one line, though it will keep working, for the most part, and a smaller number of line code and path code violations on the second line. The most recent occurrence showed that the routers both entered their respective broken states at the same time.
site A holds router A which has a t1 to router C. site A also holds the microwave equipment with the t1 to router B, though that t1 does not pass through router A.
I know this is a telco problem for the most part, but they are not totally capable of identifying a very intermittent problem like this, and I was wondering if people here had suggestions of what we should be looking for. Also, there is possibly something wrong with the cisco cards such that they get in this broken state which can be fixed by simply unplugging the t1 line.
The entire network has about 20 routers in it, with similar microwave connections and traffic. There is one other site with a similar problem (occasionally ends up with many many line code violations (few path code violations) and is reset by unplugging the t1 line), but the rest of the sites haven’t had any randomly entered bad states.
{ 2 comments… read them below or add one }
When they test the circuit, do they see any errors ? I would suggest to clear the counters on the interface and monitor if its incrementing from the time you clear it, or it increments to a high number only around the time when the intermittent problem occurs.
f you can log any syslog message that shows up to a Syslog server (you can download freeware tools like 3CDaemon or Kiwi Syslog ) that would be great. This will probably help you identify what happens at that time.
Does this event crash the routers ?
The telco can’t see any errors when they test the line.
The errors keep accumulating on the controller from the incident onward until I unplug the cable.
We don’t have a syslog set up for the routers at the moment, but I can look into setting one up temporarily.
The event doesn’t crash the routers. They seem to work normally except for the t1 errors. (They aren’t used to do much else, though.)
You must log in to post a comment.